It is generally accepted that human tumor cells contain multiple specific alterations in the cellular genome responsible for their malignant phenotype. These alterations affect the expression or function of genes that control cell growth and differentiation. For instance, typically these mutations are observed in oncogenes, or positive effectors of cellular transformation, such as ras, and in tumor suppressor genes (or recessive oncogenes) encoding negative growth regulators, the loss of function of which results in expression of a transformed phenotype, such as p53, Rb1, DCC, MCC, NF1, and WT1.
Mutations have been detected in all of the common human tumors including pancreatic and colorectal carcinomas. To date, a transforming ras gene (i.e., a mutated version of H-ras, K-ras, or N-ras encoding a protein having an altered amino acid at one of the critical positions 12, 13 and 61) is the oncogene most frequently identified in human cancer. As reviewed by Barbacid, Ann. Rev. Biochem., 56, 779-827 (1987), a ras oncogene has been observed in carcinoma of the bladder, breast, colon, kidney, liver, lung, ovary, pancreas, rectum, and stomach; in hematopoietic tumors of lymphoid and myeloid lineage; in tumors of mesenchymal origin such as fibrosarcomas and rhabdomyosarcomas; and in other tumors, including melanomas, teratocarcinomas, neuroblastomas, and gliomas. In particular, a ras mutation has been identified in greater than 90% of patients with adenocarcinoma of the pancreas, as described by Bos, Cancer Research, 49, 4682-4689 (1989).
Tumors of the pancreas are highly malignant and generally result in death. In fact, cancer of the pancreas is the fifth leading cause of cancer-related death in the United States. The presently available treatment modalities have shown little or no benefit for patients with tumors that are unresectable (i.e., regionally advanced or metastatic cancers). Similarly, for patients with localized disease that can be resected, state-of-the-art adjuvant therapy with radiation and chemotherapy has shown only modest benefits--and that at the expense of significant treatment toxicity. Over 71% of cancer patients undergoing adjuvant therapy will eventually die of recurrent disease. For these reasons, more effective treatments are currently needed for cancer, and, in particular, both for advanced as well as limited-stage pancreatic cancer.
Immunotherapy is a potentially therapeutic approach for the treatment of cancer. Immunotherapy is based on the premise that the failure of the immune system to reject spontaneously arising tumors is related to the failure of the immune system to respond appropriately to tumor antigens. In a functioning immune system, tumor antigens are processed and expressed on the cell surface in the context of major histocompatibility complex (MHC) class I and II molecules, which, in humans, also are termed "human-leukocyte associated" (HLA) molecules. When complexed to antigens, the MHC class I and II molecules are recognized by CD8.sup.+ and CD4.sup.+ T cells, respectively. This recognition generates a set of secondary cellular signals and the paracrine release of specific cytokines, or soluble so-called "biological response modifiers", that mediate interactions between cells and stimulate host defenses to fight off disease. The release of cytokines then results in the proliferation of antigen-specific T cells.
Thus, active immunotherapy involves the injection of tumor cells, typically in the vicinity of a tumor, to generate either a novel or an enhanced systemic immune response. The ability of this immunotherapeutic approach to augment a systemic T cell response against a tumor has been previously disclosed, e.g., amongst others, see International Application WO 92/05262, Fearon et al., Cell, 60, 397-403 (1990), and Dranoff et al., Proc. Natl. Acad. Sci., 90, 3539-3543 (1993). The injected tumor cells are usually altered to enhance their immunogenicity, such as by admixture with non-specific adjuvants, or by genetic modification of the cells to express cytokines, or other immune co-stimulatory molecules. The tumor cells employed can be autologous, i.e., derived from the same host as is being treated. Alternately, the tumor cells can be MHC-matched, or derived from another host having the same, or at least some of the same, MHC complex molecules.
Clinical researchers prefer the use of autologous over MHC-matched tumor cells, and vice versa, for different reasons. Namely, autologous cells are preferred since each patient's tumor expresses a unique set of tumor antigens that can differ from those found on histologically-similar, MHC-matched tumor cells from another patient, see, e.g., Kawakami et al., J. Immunol., 148, 638-643 (1992); Darrow et al., J. Immunol., 142, 3329-3335 (1989); and Hom et al., J. Immunother., 10, 153-164 (1991). Studies evaluating human melanoma antigens confirm that all the human tumor antigens identified to date are shared among at least 50% of patients' tumors--regardless of whether or not the same MHC-type is similarly shared. Use of cells from a patient's own tumor circumvents any need for matching of tumor or MHC antigens.
In comparison, MHC-matched tumor cells are preferred since the use of autologous tumor cell vaccines require that each patient be taken to surgery to obtain a sample of their tumor for vaccine production. The in vitro expansion of fresh human tumor explants necessary for the production of autologous tumor cell vaccines is labor-intensive, technically demanding, and frequently impossible for most histologic types of human tumors, even with highly specialized research facilities. Moreover, the production of a vaccine from each patient's tumor is quite expensive. There also is a substantial likelihood that after extended passage of autologous cells in culture, the antigenic composition of such cells will change relative to the primary tumor from which the cell line originated, making the cells ineffective as a vaccine. While such change is inevitable with all established cell lines, as regarding the use of autologous cells as a tumor vaccine, it will require the maintenance of freezer stocks of each initially-isolated cell line for each patient being treated using this approach.
Based on these shortcomings associated with use of autologous and MHC-matched cells as tumor vaccines, other researchers have sought alternative tumor vaccines, as reviewed by Jaffee et al., Seminars in Oncology, 22, 81-91 (1995). The recent results of Huang et al., Science, 264, 961-965 (1994), are relevant to this proposal. Namely, prior to this study, tumor vaccine strategies were based on the understanding that the vaccinating tumor cells function as the antigen presenting cells (APCs) that present the tumor antigens on their MHC class I and II molecules, and directly activate the T cell arm of the immune response. In contrast, the results of Huang et al. indicate that the professional APCs of the host rather than the vaccinating tumor cells prime the T cell arm of the immune response. The tumor vaccine cells secrete a cytokine such as GM-CSF and recruit to the region of the tumor bone marrow-derived APCs. The bone marrow-derived APCs take up the whole cellular protein of the tumor for processing, and then present the antigenic peptide(s) on their MHC class I and II molecules. In this fashion, the APCs prime both the CD4.sup.+ and the CD8.sup.+ T cell arms of the immune system, resulting in the generation of a systemic antitumor immune response that is specific for the antigenic epitopes of the host tumor. These results suggest that it may not be necessary to use autologous or MHC-matched tumor cells in cancer treatment.
Other results suggest that the transfer of allogeneic MHC genes (i.e., genes from a genetically dissimilar individual of the same species) can enhance tumor immunogenicity. Specifically, in certain cases, the rejection of tumors expressing allogeneic MHC class I molecules resulted in enhanced systemic immune responses against subsequent challenge with the unmodified parental tumor, as reviewed in Jaffee et al., supra, and Huang et al., supra. This appears to represent an example of the general phenomenon described as "xenogenization" by Itaya et al., Cancer Res., 47, 3136-3140 (1987), wherein tumor vaccine potency is enhanced by introducing genes into the tumor cell that code for foreign antigens.
Thus, there remains a need for a method of treating cancer, in particular, a method of treating pancreatic cancer, which does not rely on use of autologous or MHC-matched tumor cells, and that avoids the difficulties and shortcomings associated with such use. The present invention provides such a method, as well as components necessary for effectuating the method. These and other objects and advantages of the present invention, as well as additional inventive features, will be apparent from the description of the invention set forth herein.